One even-more-last note on window glass:
 
Benjamin-
 
Quick and easy ways to identify window glass types:
 
The three technological types of window glass do not have any sharp cut-off dates, but identifying them can often be broadly useful for dating sites and deposits in relation to the periods in which they were primary forms of manufacture: 
 
"Crown glass," or the type made in flat sheets manipulated with an iron pontil (Diderot plates 235-248) seems to have been the primary method used (at least in New York) until approximately the 1820s-1830s. It tends to be very variable in color, ranging from darker green to aqua and even blue. Because the "table," or sheet of glass, was thinner towards the edge and thicker in the middle, it is quite variable in thickness, often visibly on the same sherd.
 
"Cylinder glass," produced industrially in a manner derived from what Diderot called "broad glass" (Plates 249-254) was made from a large cylindrical bottle by cutting the ends off with a hot wire and slitting it down one side, then placing it in an annealing oven to make it lie flat. This method is equally ancient, but was most common in the U.S. from the early 19th century to the early 20th century ca. 1930. It tends to be lighter aqua to clear in color, thicker than crown glass, and relatively uniform in thickness. This method was rendered obsolete by the adoption of efficient methods of producing cheap plate glass, because plate glass does not distort the image seen through it.
 
These two earlier types can easily be distinguished from plate glass in the lab by the fact that images viewed through even tiny pieces look rippled or distorted when the piece is moved back and forth.
 
I have found window glass useful only as a source of very general information about dates, but beyond telling you very roughly when a structure was built (e.g. before ~1830) and how long it was occupied, it usually doesn't tell you much. 
 
More technical methods of analysis might reveal more information such as sources, but in the context of CRM, as you pointed out, we usually don't have that luxury. Frankly, the responses to your query provided more information of use to me than the window glass itself ever has. Now, of course, my very next project will undoubtedly make a liar out of me.
 
 
Marty Pickands
New York State Museum